Ground segmentation , point cloud clustering based on CVC(Curved Voxel Clustering) and polygonal box(classify vehicles simply by length,width and height) 本项目使用设置地面坡度阈值的方法，滤除地面点，使用三维弯曲体素聚类法完成点云的聚类，包围盒参数由Apollo库的common_lib和object_builders_lib 得出,代码还有一些中文注释。 运行步骤：
git clone https://github.com/C-Xingyu/my_detection.git
cd my_detection
catkin_make -DCATKIN_WHITELIST_PACKAGES="common_lib"
catkin_make -DCATKIN_WHITELIST_PACKAGES="object_builders_lib"
catkin_make -DCATKIN_WHITELIST_PACKAGES="my_detection"
接下来配置launch文件（velodyne_points.launch） 运行
roslaunch my_detection velodyne_points.launch
